INDEX

Explanations

phrases related to describing events or incidents as witnessed by individuals

oai_token-act-pair · gpt-3.5-turbo

phrases related to witnessing or reporting events

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.6.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Britann

-0.73

ãĥ¯

-0.68

çķ

-0.68

Correction

-0.65

OTUS

-0.65

ahead

-0.65

boot

-0.63

 wink

-0.62

cade

-0.62

 isEnabled

-0.62

POSITIVE LOGITS

ewitness

1.07

 testimonies

0.90

 anecd

0.87

 harrowing

0.85

 testim

0.80

 abuse

0.76

 experiencing

0.76

 suffered

0.76

 complaints

0.75

 complain

0.74

Activations Density 0.424%

phrases related to describing events or incidents as witnessed by individuals

phrases related to witnessing or reporting events

No Comments

No Known Activations